P0613 - Utility of Length of Stay and Direct Cost in Evaluating for Disparities in Pediatric Inpatient Care on a General Medical/Surgical Ward

BACKGROUND/

Purpose:
Numerous studies have described disparities in healthcare utilization, diagnosis, management, and outcomes of pediatric medical and surgical conditions according to patient demographic factors. However, disparities in care may also depend on multiple local factors, and published trends may not apply to individual hospitals and patient populations. Additionally, methods for data collection and available metrics may vary substantially between health systems. The purpose of this study was to retrospectively assess our local tertiary care institution for disparities in outcomes among inpatients discharged from a pediatric medical/surgical ward.

Methods:
Over 2 fiscal years, inpatient data were obtained from hospital administrative and billing systems and a healthcare performance improvement services company, and the 10 most frequent primary diagnoses were determined. For each of these diagnoses, we used 2-tailed student’s t-test or one-way ANOVA, as appropriate, to compare (1) observed hospital length of stay (LOS) and (2) inpatient direct cost, by gender, race, ethnicity, and primary payor. A rejection level of p<0.05 was considered significant.

Results:
There were 2795 inpatient discharges from the ward of interest during the period under study, with bronchiolitis and adenotonsillar hypertrophy the most predominant among the top 10 primary diagnoses (Table 1). For the primary diagnosis of dehydration, discharges involving private preferred provider organizations had significantly higher observed LOS and direct cost (p<0.01 and p=0.03, respectively), although there were relatively few visits in this group. For all other primary diagnoses, no significant differences were found comparing LOS or direct costs by gender, race, ethnicity, or primary payor.

Conclusion:
Despite previously published disparities, we were unable to find consistent differences between demographic groups for LOS or direct cost according to primary diagnoses on our general inpatient ward. Our methodology may provide a starting framework for other institutions to assess for local disparities in outcomes among pediatric inpatients. Although our results are reassuring, our analyses were limited by the lack of risk adjustment, and our approach also excluded bedded outpatient and observation visits. Outcomes other than absolute LOS or direct cost, such as comparing indices of observed values to those expected based on national benchmarks, may represent better metrics to assess for disparities.